simba (version 0.3-4)

sim.tmp: Calculate binary similarity in time

Description

The function applies one of 56 similarity measures for binary data to calculate compositional similarity of plots between time steps.

Usage

sim.tmp(x, y, method = "soer", normalize = FALSE, adjust = TRUE, ...)

Arguments

x

Vegetation data, either as matrix with rows = plots and columns = species, or as data.frame with first three columns representing plots, species and occurence information respectively. All

y

Same as x for time-step two.

method

One of 42 similarity measures for binary data. The function uses the same indices as sim. See details there. Per default soerensen similarity is calculated.

normalize

Logical value indicating whether the values for a, b and c which are calculated in the process should be normalized to 100% (per row, which means per plot comparison). If normalize = TRUE an asymmetric index must

adjust

Do not change the default behaviour (TRUE) unless you know what you do. Would spare some calculation time if set to FALSE, when your species data do not need adjustment, which means that in both or all time steps, there are exactly the same species and th

...

Other arguments to sim

Value

• Returns a named vector with the similarities for each site between time steps for each plot.

encoding

UTF-8

Details

If you compare species data among time steps there will be most likely different numbers of species (and often also different numbers of plots for which information is available). The function takes care of this and you can give any species matrices you want. If one plot is the same, it will calculate what changed on this plot. There will be an error message if no plot is shared. The function relies on plot and species names!! As in a database - they must be unique!!

References

See references in sim

See Also

See Also as sim (where you can find a much more elaborate see also section as well).

Examples

data(bernina)
## load included data
data(bernina)

## how species changed occurrence between first recording
## and last recording?
# construct a species matrix that only contains the species
# that occurred on the summits at the first recording
first <- veg[summits$year=="1907",]
first <- first[,colSums(first)>0]
# make right summit names
row.names(first) <- as.character(summits$summit[summits$year=="1907"])
# construct a species matrix that only contains the species
# that occurred on the summits at the last recording
last <- veg[summits$year=="2003",]
last <- last[,colSums(last)>0]
# make right summit names
row.names(last) <- as.character(summits$summit[summits$year=="2003"])
# calculate similarity between time steps
sim.tmp(first, last)